US 11,734,517 B1
Systems and methods for measuring automatability of report generation using a natural language generation system
Hanna Abi-Akl, Paris (FR); Hugues Sézille de Mazancourt, Thiais (FR); Dominique Mariko, Paris (FR); and Valentin Pierart, Courbevoie (FR)
Assigned to YSEOP SA, Lyons (FR)
Filed by YSEOP SA, Lyons (FR)
Filed on Feb. 28, 2022, as Appl. No. 17/683,060.
Claims priority of provisional application 63/158,454, filed on Mar. 9, 2021.
Int. Cl. G06F 17/00 (2019.01); G06F 40/30 (2020.01); G06F 40/117 (2020.01); G06F 40/169 (2020.01); G06F 40/284 (2020.01)
CPC G06F 40/30 (2020.01) [G06F 40/117 (2020.01); G06F 40/169 (2020.01); G06F 40/284 (2020.01)] 19 Claims
OG exemplary drawing
 
1. A method for determining a degree to which a document can be generated using a natural language generation (NLG) system, the NLG system being configured to generate natural language text using semantic objects, the method comprising:
using at least one computer hardware processor to perform:
(A) obtaining a document comprising text segments;
(B) determining a degree to which at least some of the text segments can be generated using the NLG system and at least some of the semantic objects;
(C) generating a report indicating the degree to which the at least some of the text segments can be generated using the NLG system and the at least some of the semantic objects; and
(D) outputting the report,
wherein the at least some of the text segments include a first text segment, and wherein (B) comprises:
generating a first annotated representation for the first text segment; and
determining, for each semantic object of one or more of the at least some of the semantic objects, a degree to which the first text segment can be generated by the NLG system using the semantic object, the determining comprising:
accessing a plurality of annotated representations for text segments associated with the semantic object;
determining measures of similarity between the first annotated representation and the plurality of annotated representations; and
determining, using the measures of similarity, the degree to which the first text segment can be generated by the NLG system using the semantic object.